Effect of Grain Size Distribution on California Bearing Ratio (CBR) and Modified Proctor Parameters for Granular Materials

Abstract

The California bearing ratio (CBR) and modified proctor parameters (maximum dry unit weight γd(max)γd(max) and optimum moisture content woptwopt) are valuable indicators of the compaction quality of subgrades, embankments and granular fills. In the engineering practice, correlations of these variables with granulometric properties of the soil are required, especially since testing for these variables can be time-consuming when a large number of samples are analyzed. In this work, 20 different granular materials with varying grain size distributions were prepared and tested. Their grain size distribution properties and their parameters CBR, γd(max)γd(max) and woptwopt were determined. These results were analyzed along with a compilation of 77 additional experimental results on granular materials reported in the literature. The influence of some granulometric properties on the parameters CBR, γd(max)γd(max) and woptwopt was statistically examined, and some correlations were proposed for these variables. Subsequently, it was demonstrated that the proposed correlations show better accuracy tother reported correlations in the literature. Finally, this work ends with some concluding remarks.

Surrogate models to predict maximum dry unit weight, optimum moisture content and California bearing ratio form grain size distribution curve

“…2-The model proposed in this paper to predict the optimum moisture contains prediction (Equation 7) demonstrated better predictions than the model proposed by Duque et al (2020). This model scored 𝐺𝐺𝐴𝐴𝐶𝐶, 𝐶𝐶𝐺𝐺𝑆𝑆𝐶𝐶, 𝜇𝜇, 𝑎𝑎20 − 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 and 𝐶𝐶 of 0.76, 0.98, 1.01, 0.96, and 0.89, respectively for the training dataset and 0.49, 0.65, 1.00, 1.00 and 0.96, respectively for the validation dataset (see Table 6 for detailed comparison).…”

“…In addition, the 𝑎𝑎20 − 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 values show that the new model provides 100% predictions within an error range of ±20% for the validation dataset, while Duque et al (2020)…”

“…Tables 5 compares the statistical performance of Equation 6with the models presented in Table 1 ( Gurtug et al, 2004;Gurtug et al, 2018;Duque et al, 2020) 2020) is more accurate than the model developed by Gurtug et al (2004) and Gurtug et al (2018), where Gurtug et al (2004) model scored very low coefficient of correlation and did not predict any point within error range of ±20% and the model of Gurtug et al (2018) scored lower coefficient of correlation for both dataset, higher error for both dataset, and lower a20-index for the validation data compared with the new model.…”

“…Table 6 assesses the performance of Equation 7 against the model proposed by Duque et al (2020) to predict the optimum moisture content. The scored statistical indicators also demonstrate that the new model is better than the available one, and for both datasets.…”

Stabilization of Coal Mine Waste and Its Practical Application as a Road Subgrade

“…Effective size (D 10 ) is 0.0047, Coefficient of gradation (C c ) is 0.37, Uniformity coefficient (C u ) is 13.33, and Sorting coefficient (S o ) is 2.95. As much the D 10 , C c and C u increases the CBR value increases [17]. The particle size is good to be used as a subgrade material.…”